Assessing the shear behavior of steel fiber reinforced concrete beams corroded under chloride attacks

Steel corrosion affects the failure mechanism of deteriorated reinforced concrete (RC) beams. Meanwhile, there is a lack of research on the shear behavior of corroded RC beams, particularly corroded steel fiber reinforced concrete (SFRC) beams. This paper investigates the shear behavior of corroded SFRC beams with a 1.5 shear span-to-depth ratio. All beam specimens included steel fibers with 50 kg/m3.

Trang 1

Journal of Science and Technology in Civil Engineering, HUCE (NUCE), 2022, 16 (3): 97–110

ASSESSING THE SHEAR BEHAVIOR OF STEEL FIBER REINFORCED CONCRETE BEAMS CORRODED UNDER

CHLORIDE ATTACKS

Nguyen Thi Thanh Thaoa, Tran Phi Son Tunga, Nguyen Duc Nhana, Nguyen Ngoc Tana,∗

a

Faculty of Building and Industrial Construction, Hanoi University of Civil Engineering,

55 Giai Phong road, Hai Ba Trung district, Hanoi, Vietnam

Article history:

Received 12/4/2022, Revised 10/5/2022, Accepted 18/5/2022

Abstract

Steel corrosion affects the failure mechanism of deteriorated reinforced concrete (RC) beams Meanwhile, there

is a lack of research on the shear behavior of corroded RC beams, particularly corroded steel fiber reinforced concrete (SFRC) beams This paper investigates the shear behavior of corroded SFRC beams with a 1.5 shear span-to-depth ratio All beam specimens included steel fibers with 50 kg/m3 In particular, tensile longitudinal reinforcements were subjected to a corrosion degree of 16.4%, while stirrups were subjected to an approxi-mately 24.1% corrosion degree These results are compared to those obtained on a non-corroded beam The obtained results from the four-point loading tests show that the corroded SFRC beams preserve a softening behavior as opposed to the sudden shear failure of RC beams without steel fibers.

Keywords:shear behavior; steel fiber reinforced concrete; reinforced concrete beam; steel corrosion.

1 Introduction

Nowadays, there are undoubtedly several benefits of steel fiber reinforced concrete (SFRC) One

of the primary positives of SFRC is that it improves the mechanical properties of traditional concrete, particularly the tensile strength [1,2] This means that the bridging actions of steel fibers on cracks

in SFRC during loading could help to reduce crack width Moreover, one of the advantages of SFRC

is crack resistance in enhanced post-crack strength

The addition of steel fibers to concrete that may be used to enhance the mechanical properties has been studied experimentally Nguyen et al [3] found the following results after testing sample groups constructed of SFRC with varying steel fiber contents: (i) When the steel fiber content was raised, the tensile strength of SFRC grew from 8 to 97%; (ii) Unlike traditional concrete, the addition

of steel fibers enhanced the ductility response Likewise, another research work conducted by Bui et

al [4] presented that when added steel fibers splitting tensile strength and flexural tensile strength significantly increased by up to 228% and 145%, respectively Steel fiber addition to beam structures exhibited more ductility than ordinary concrete in the post-peak stage

On the other hand, shear failure is a critical issue in the RC beams because of the brittle behavior [5] Shear strength is generally predicted using the geometric theory of the shear resistant mechanism

∗

Corresponding author E-mail address:tannn@huce.edu.vn (Tan, N N.)

97

Trang 2

Thao, N T T., et al / Journal of Science and Technology in Civil Engineering

According to the design code ACI 318-19 [6], if a ratio of shear span-to-effective depth (a/d) is less than 2, beams are classified as deep beams with an arch action for the shear resistant mechanism However, a beam action mechanism is a way of load transfer in shear for beams having an a/d ra-tio of greater than 2.0 Meanwhile, current building construcra-tions in Vietnam have been frequently subjected to harsh environmental conditions, which exacerbate the corrosion of RC structures Steel corrosion is a severe problem that occurs in the loss of bond between concrete and steel reinforce-ment resulting in a decrease in both flexural and shear capacity of RC structures [7 9] Soltani et al [7] reported that flexural and shear strengths of beam specimens were reduced to 80% in corrosive environments Besides, previous studies have shown that the failure modes of corroded RC structures can change into brittle failure due to a higher degree of corrosion, especially in shear performance [10–14] It was shown in the study of Nguyen et al [15] that steel corrosion may modify the shear transferring mechanism of corroded RC beams For example, beam specimens with an a/d ratio less than 2.0 might have the load transferring mode changing from the combination of beam action and arch action to mainly arch action However, beams with an a/d ratio greater than 2.0 may fail in diagonal tension failure by beam action

Numerous researches have investigated the influence of steel fibers on the ability of concrete structures in flexural and shear capacity based on the improved ductile performance in SFRC [16–

21] Particularly, Bui et al [4] further claimed that steel fibers with a high-volume fraction of greater than 1.2% were capable of replacing stirrups to ensure shear capacity because of the similar behavior

to traditional RC beams Additionally, Kwak et al [19] investigated that beams with an a/d ratio of 2.0 failed in a combination of shear and flexure, whereas beams with a higher a/d ratio only failed

by flexure However, Biolzi et al [21] argued that beams with an a/d ratio of 1.5 still failed by arch action with a ductile post-peak part

The steel fibers in concrete not only improve the overall flexural and shear resistance but also is

an effective method for corrosion resistance Furthermore, steel fibers could restrain the propagation

of corrosion-induced cracks due to the restricted migration and diffusion transport capabilities of concrete [22, 23] Taqi et al [24] carried out experimental works that discussed the influence of corrosion on the shear behavior of SFRC beams having a 2.8 a/d ratio with or without pre-corroded steel fiber They claimed that increasing the steel fiber content causes the failure mode to change from shear to flexural failure

Apart from the advantages of steel fibers on concrete properties and structural behavior, the cor-rosion resistance of SFRC is rarely considered The present experiment in this paper focuses on the shear performance of one control and two corroded SFRC beams with a 0.6% volume fraction of steel fiber hooked-end type (corresponding to 50 kg/m3) First, the corroded beam specimens were taken to an accelerated corrosion test with 16.4% and 24.1% average degrees of corrosion for ten-sile longitudinal reinforcement and stirrups Then, after determining the material properties, three beam specimens constructed entirely of steel fiber reinforced concrete were subjected to a four-point loading test Finally, the findings on the role of steel fibers in the shear behavior of SFRC beams by chloride attacks through the experimental study were discussed

2 Experimental program

2.1 Materials

Table1shows the designed mixture of SFRC used As with ordinary concrete, Portland cement PCB40 was used as the binder, while river sand and crushed stone were used for fine and coarse

98

Trang 3

Journal of Science and Technology in Civil Engineering HUCE 2022 ISSN 1859-2996

3

They claimed that increasing the steel fiber content causes the failure mode to change from shear to flexural failure

Apart from the advantages of steel fibers on concrete properties and structural behavior, the corrosion resistance of SFRC is rarely considered The present experiment

in this paper focuses on the shear performance of one control and two corroded SFRC beams with a 0.6% volume fraction of steel fiber hooked-end type (corresponding to 50 kg/m3) First, the corroded beam specimens were taken to an accelerated corrosion test with 16.4% and 24.1% average degrees of corrosion for tensile longitudinal reinforcement and stirrups Then, after determining the material properties, three beam specimens constructed entirely of steel fiber reinforced concrete were subjected to a four-point loading test Finally, the findings on the role of steel fibers in the shear behavior of SFRC beams by chloride attacks through the experimental study were discussed

2 Experimental program

2.1 Materials

Table 1 shows the designed mixture proportions of SFRC As with ordinary concrete, Portland cement PCB40 was used as the binder, while river sand and crushed stone were used for fine and coarse aggregates The steel fiber with a content of 50 kg/m3 by mass was used in the SFRC 50 mix These steel fibers are manufactured from stainless steel provided by Bekaert The experimental works used Dramix 3D 65/35BG steel fibers deformed in 3D with hooked ends to increase anchoring in the concrete matrix, as shown in Fig 1 In particular, the elastic modulus of the fiber was around 210 GPa, and the tensile strength was 1345 MPa The aspect ratio of fiber length to diameter

(l f /d f ) was 65 for defining the properties of the fibers (corresponding to a length l f of

35mm and a diameter d f of 0.55mm)

Fig 1 Dramix 3D 65/35BG hooked-end steel fibers The cubes with the dimensions of 150x150x150 mm were cast, cured in indoor environmental conditions of the laboratory, and tested to determine compressive strength As a result, the mean compressive strength of a set of three SFRC cubes at 28 days was 49.8 MPa, as shown in Table 2, equivalent to the concrete C40/50, as defined

Figure 1 Dramix 3D 65/35BG hooked-end

steel fibers

aggregates The steel fibers with a content of

50 kg/m3by mass were used in the SFRC 50 mix

These steel fibers are manufactured from

stain-less steel provided by Bekaert The experimental

works used Dramix 3D 65/35BG steel fibers

de-formed in 3D with hooked ends to increase

an-choring in the concrete matrix, as shown in Fig

1 In particular, the elastic modulus of the fiber

was around 210 GPa, and the tensile strength was

1345 MPa The aspect ratio of fiber length to

di-ameter (lf/df) was 65 for defining the properties

of the fibers (corresponding to a length lf of 35

mm and a diameter df of 0.55 mm)

Table 1 Concrete mix

(kg/m3)

Fine aggregates (kg/m3)

Coarse aggregates (kg/m3)

Water (liter/m3)

Ratio W/C

Steel fiber (kg/m3)

The cubes with the dimensions of 150×150×150 mm were cast, cured in indoor environmental

conditions of the laboratory, and tested to determine compressive strength As a result, the mean

compressive strength of a set of three SFRC cubes at 28 days was 49.8 MPa, as shown in Table 2,

equivalent to the concrete C40/50, as defined in Eurocode 2 (EC2) [25]

Table 2 Compressive strength of SFRC specimens

Sample

Maximum compressive load (kN)

Compressive strength (MPa)

Mean compressive strength (MPa)

Standard deviation (MPa)

Coefficient

of variation (%)

In this study, beam specimens used longitudinal reinforcements of 10-mm and 12-mm diameter

and stirrups of 6-mm diameter, as illustrated in Fig 2 There are three sets of steel rebars and each

of which has three specimens manufactured from the same strength grades of steel The yield and

ultimate tensile strengths of steel reinforcements were determined by the tension test The obtained

results are synthesized in Table3

2.2 SFRC beam specimens

According to the SFRC mixture, three beam specimens were conducted The first beam, named

B1.1-NC, is the non-corroded SFRC beam as the control beam, while two beams, denoted B2.1-C

99

Trang 4

Table 3 Tensile strength of steel reinforcements

Sample

Diameter/

steel

type

Yield tensile load (kN)

Ultimate tensile load (kN)

Yield tensile strength (MPa)

Mean yield tensile strength (MPa)

Ultimate tensile strength (MPa)

Mean ultimate tensile strength (MPa)

CB300-V

374.3

534.9

543.2

CB300-V

366.9

563.2

565.4

CB240-T

331.1

519.1

514.8

and B3.1-C, are subjected to the accelerated corrosion test Fig.2illustrates the dimension and layout

of the beam specimens In more detail, the beam specimens have a width of 150 mm, a height of 200

mm, and a length of 1100 mm Three beams were reinforced with two ϕ10 mm and two ϕ12 mm steel rebars in the top and bottom layers, respectively Additionally, all beam specimens were installed with ϕ6 mm stirrups with a regular spacing of 150 mm The concrete cover is 40 mm in thickness according

to the Vietnamese standard TCVN 9346:2012 [26] for RC structures in aggressive conditions

2.3 Accelerated electrochemical corrosion test

Analyzing the load-carrying capacity of corroded RC beams under normal service conditions is more challenging, depending on the corrosion rate [27] In this study, an accelerated electrochemi-cal corrosion test was conducted to produce experimental specimens that exhibit the same corrosion behavior as in reality in a shorter period Two beam specimens were immersed in a 3.5% sodium solution (NaCl) for 48 hours for concrete total saturation The accelerated electrochemical corrosion test on the beam specimens is shown in Fig.3 Simultaneously, chloride ions diffused into the corro-sion test specimens from the electrolyte solution Each transformer was connected to the longitudinal rebars at the bottom layer of corroded beams Each longitudinal rebar received an electrical current of 1A maintained constant during the corrosion test process Finally, corrosion testing of SFRC beams would be completed after 576 hours continuously, which was estimated based on Faraday’s law [28]

5

According to the SFRC mixture, three beam specimens were conducted The first

beam, named B1.1-NC, is the non-corroded SFRC beam as the control beam, while two

beams, denoted B2.1-C and B3.1-C, are subjected to the accelerated corrosion test Fig

2 illustrates the dimension and layout of the beam specimens In more detail, the beam

specimens have a width of 150 mm, a height of 200 mm, and a length of 1100 mm

Three beams were reinforced with two ϕ10 mm and two ϕ12 mm steel rebars in the top

and bottom layers, respectively Additionally, all beam specimens were installed with

ϕ6 mm stirrups with a regular spacing of 150 mm The concrete cover is 40 mm in

thickness according to the Vietnamese standard TCVN 9346:2012 [27] for RC

structures in aggressive conditions[26]

Fig 2 Detailed layout of beam specimens

(a) Testing diagram (b) SFRC beams in the solution Fig 3 Accelerated corrosion test for corroded SFRC beams

Analyzing the load-carrying capacity of corroded RC beams under normal service

conditions is more challenging, depending on the corrosion rate [26] In this study, an

accelerated electrochemical corrosion test was conducted to produce experimental

specimens that exhibit the same corrosion behavior as in reality in a shorter period Two

beam specimens were immersed in a 3.5% sodium solution (NaCl) for 48 hours for

concrete total saturation The accelerated electrochemical corrosion test on the beam

specimens is shown in Fig 3 Simultaneously, chloride ions diffused into the corrosion

test specimens from the electrolyte solution Each transformer was connected to the

longitudinal rebars at the bottom layer of corroded beams Each longitudinal rebar

received an electrical current of 1A maintained constant during the corrosion test

1-1

Figure 2 Detailed layout of beam specimens

100

Trang 5

5

According to the SFRC mixture, three beam specimens were conducted The first beam, named B1.1-NC, is the non-corroded SFRC beam as the control beam, while two beams, denoted B2.1-C and B3.1-C, are subjected to the accelerated corrosion test Fig

2 illustrates the dimension and layout of the beam specimens In more detail, the beam specimens have a width of 150 mm, a height of 200 mm, and a length of 1100 mm Three beams were reinforced with two ϕ10 mm and two ϕ12 mm steel rebars in the top and bottom layers, respectively Additionally, all beam specimens were installed with ϕ6 mm stirrups with a regular spacing of 150 mm The concrete cover is 40 mm in thickness according to the Vietnamese standard TCVN 9346:2012 [27] for RC structures in aggressive conditions[26]

Fig 3 Accelerated corrosion test for corroded SFRC beams Analyzing the load-carrying capacity of corroded RC beams under normal service conditions is more challenging, depending on the corrosion rate [26] In this study, an accelerated electrochemical corrosion test was conducted to produce experimental

beam specimens were immersed in a 3.5% sodium solution (NaCl) for 48 hours for concrete total saturation The accelerated electrochemical corrosion test on the beam specimens is shown in Fig 3 Simultaneously, chloride ions diffused into the corrosion test specimens from the electrolyte solution Each transformer was connected to the longitudinal rebars at the bottom layer of corroded beams Each longitudinal rebar received an electrical current of 1A maintained constant during the corrosion test

1-1

(a) Testing diagram Journal of Science and Technology in Civil Engineering HUCE 2022 ISSN 1859-2996

5

According to the SFRC mixture, three beam specimens were conducted The first beam, named B1.1-NC, is the non-corroded SFRC beam as the control beam, while two beams, denoted B2.1-C and B3.1-C, are subjected to the accelerated corrosion test Fig

2 illustrates the dimension and layout of the beam specimens In more detail, the beam specimens have a width of 150 mm, a height of 200 mm, and a length of 1100 mm Three beams were reinforced with two ϕ10 mm and two ϕ12 mm steel rebars in the top and bottom layers, respectively Additionally, all beam specimens were installed with ϕ6 mm stirrups with a regular spacing of 150 mm The concrete cover is 40 mm in thickness according to the Vietnamese standard TCVN 9346:2012 [27] for RC structures in aggressive conditions[26]

(a) Testing diagram (b) SFRC beams in the solution Fig 3 Accelerated corrosion test for corroded SFRC beams

Analyzing the load-carrying capacity of corroded RC beams under normal service conditions is more challenging, depending on the corrosion rate [26] In this study, an accelerated electrochemical corrosion test was conducted to produce experimental specimens that exhibit the same corrosion behavior as in reality in a shorter period Two beam specimens were immersed in a 3.5% sodium solution (NaCl) for 48 hours for concrete total saturation The accelerated electrochemical corrosion test on the beam specimens is shown in Fig 3 Simultaneously, chloride ions diffused into the corrosion test specimens from the electrolyte solution Each transformer was connected to the longitudinal rebars at the bottom layer of corroded beams Each longitudinal rebar received an electrical current of 1A maintained constant during the corrosion test

1-1

(b) SFRC beams in the solution Figure 3 Accelerated corrosion test for corroded SFRC beams

2.4 Four-point loading test

Following the corrosion test, an experimental program was carried out on three beam specimens to assess the mechanical behavior of non-corroded and corroded SFRC beams based on several param-eters, such as the load-displacement curves, the relationship between load and crack width, cracking pattern, and failure mechanism

6

process Finally, corrosion testing of SFRC beams would be completed after 576 hours

continuously, which was estimated based on Faraday's law [28]

Following the corrosion test, an experimental program was carried out on three

beam specimens to assess the mechanical behavior of non-corroded and corroded SFRC

beams based on several parameters, such as the load-displacement curves, the

relationship between load and crack width, cracking pattern, and failure mechanism

(a) Control beam (b) Corroded beam

Fig 4 Experimental set-up for (a) the non-corroded beam, (b) the corroded beam

Fig 5 Four-point loading test configuration Figs 4 and 5 depict an experimental program of three beams on a four-point

loading test The span of beams between the supports was 900 mm The load was

distributed on beams in two loading application points by a hydraulic jack at a

controlled speed The distance between the two loading points is 450 mm, and the

distance between the support and the loading point is 225 mm So, the shear

span-to-depth ratio (a/d) of experimental beams is 1.5 Six linear variable differential

transformers (LVDT) were arranged to measure vertical displacements and crack width

For the vertical displacement, devices I1 and I3 were placed at the position of two

supports, device I2 was located on the bottom face and at the middle span, and device

ITH was located at the beam's neutral axis For the crack mouth opening displacement

(a) Control beam Journal of Science and Technology in Civil Engineering HUCE 2022 ISSN 1859-2996

6

process Finally, corrosion testing of SFRC beams would be completed after 576 hours continuously, which was estimated based on Faraday's law [28]

Following the corrosion test, an experimental program was carried out on three beam specimens to assess the mechanical behavior of non-corroded and corroded SFRC beams based on several parameters, such as the load-displacement curves, the relationship between load and crack width, cracking pattern, and failure mechanism

(a) Control beam (b) Corroded beam Fig 4 Experimental set-up for (a) the non-corroded beam, (b) the corroded beam

Fig 5 Four-point loading test configuration Figs 4 and 5 depict an experimental program of three beams on a four-point loading test The span of beams between the supports was 900 mm The load was distributed on beams in two loading application points by a hydraulic jack at a controlled speed The distance between the two loading points is 450 mm, and the distance between the support and the loading point is 225 mm So, the shear

transformers (LVDT) were arranged to measure vertical displacements and crack width For the vertical displacement, devices I1 and I3 were placed at the position of two supports, device I2 was located on the bottom face and at the middle span, and device

(b) Corroded beam Figure 4 Experimental set-up for (a) the non-corroded beam, (b) the corroded beam

Figs.4 and5 depict an experimental program of three beams on a four-point loading test The span of beams between the supports was 900 mm The load was distributed on beams in two loading application points by a hydraulic jack at a controlled speed The distance between the two loading points is 450 mm, and the distance between the support and the loading point is 225 mm So, the shear span-to-depth ratio (a/d) of experimental beams is 1.5 Six linear variable differential trans-formers (LVDT) were arranged to measure vertical displacements and crack width For the vertical displacement, devices I1and I3were placed at the position of two supports, device I2was located on the bottom face and at the middle span, and device ITH was located at the beam’s neutral axis For the crack mouth opening displacement (CMOD) of shear cracks, two devices denoted CR1 and CR2 were perpendicularly located on the diagonal lines between loading point centers and supports, and a 2-cm distance for the bottom face All testing devices were connected to a data logger TDS-530 and

a laptop computer to automatically record the data

101

Trang 6

6

process Finally, corrosion testing of SFRC beams would be completed after 576 hours continuously, which was estimated based on Faraday's law [28]

Following the corrosion test, an experimental program was carried out on three beam specimens to assess the mechanical behavior of non-corroded and corroded SFRC beams based on several parameters, such as the load-displacement curves, the relationship between load and crack width, cracking pattern, and failure mechanism

Fig 4 Experimental set-up for (a) the non-corroded beam, (b) the corroded beam

Fig 5 Four-point loading test configuration Figs 4 and 5 depict an experimental program of three beams on a four-point loading test The span of beams between the supports was 900 mm The load was distributed on beams in two loading application points by a hydraulic jack at a controlled speed The distance between the two loading points is 450 mm, and the distance between the support and the loading point is 225 mm So, the shear

transformers (LVDT) were arranged to measure vertical displacements and crack width For the vertical displacement, devices I1 and I3 were placed at the position of two supports, device I2 was located on the bottom face and at the middle span, and device

Figure 5 Four-point loading test configuration

3 Experimental results

This study focuses on the shear behavior of non-corroded and corroded SFRC beams Therefore, four-point loading tests were conducted on all specimens until failure after two beams (B2.1-C and B3.1-C) were subjected to the corrosion process The results of the accelerated corrosion process and the four-point loading tests are included in this section

3.1 Actual corrosion degree

In this study, the corrosion degree of each steel reinforcement, denoted c (%), is determined by

Eq (1) Then, the average degree of corrosion, denoted cm(%), is calculated for all steel reinforce-ments of the same type

c(%)= mo− m

where: mo(in g) is the original weight of the steel rebar before corrosion, which was measured before casting the beam specimens; m (in g) is the final weight of the rebar after corrosion, which was measured after cleaning;∆m is the mass loss of the steel rebar

After completing the four-point loading test, all longitudinal reinforcements and stirrups were removed to determine the mass loss of corroded rebars When the surrounding concrete part of the corroded beams was entirely demolished, corroded rebars were extracted carefully After an accel-erated electrochemical corrosion process, longitudinal rebars and stirrups exhibited corrosion attack, but the stainless steel fibers were maintained corrosion-free, as shown in Fig.6 In the following step, longitudinal rebars and stirrups were cleaned using HCl solution according to ASTM G1-03 [29] The residual weights of longitudinal rebars and stirrups were measured, and their corrosion degrees are reported in Table 4 Four longitudinal rebars in each beam specimen are numbered from 1 to 4 with the designation R, while eight stirrups are denoted with the designation S and numbers ranging from 1 to 8 As a result of mass loss, the corrosion degrees of tensile longitudinal rebars and stirrups were 16.4% and 24.1% on average, respectively Meanwhile, the corrosion degree of compressive longitudinal rebars at the top layer of the beam specimens have small corrosion degrees ranging from 5.4% to 7.0% on average In this study, the corrosion degree of corroded SFRC beams is considered equivalent to that of tension longitudinal reinforcement at the bottom layer

102

Trang 7

8

the bottom layer

Fig 6 Stage of a corroded SFRC beam when extracting rebars

Table 4 Corrosion results in longitudinal reinforcements and stirrups

Beam Steel sample mo (g) m (g) ∆m (g) c (%) cm (%)

B2.1-C

R2-1 894.7 744.5 150.2 16.8

16.5 R2-2 905.5 758.5 147.0 16.2

7.0

25.8

B3.1-C

R3-1 894.6 774.0 120.6 13.5

16.3 R3-2 899.1 726.5 172.6 19.2

5.4

22.5

3.2 Mechanical behavior of the control SFRC beam

The mechanical behavior of the control beam B1.1-NC is analyzed based on the load-displacement curves, as shown in Fig 7 The solid line shows the mid-span

displacement measured at the bottom face (denoted fb), while the dotted line presents

Figure 6 Stage of a corroded SFRC beam when extracting rebars Table 4 Corrosion results in longitudinal reinforcements and stirrups

B2.1-C

16.5

7.0

25.8

B3.1-C

16.3

5.4

22.5

3.2 Mechanical behavior of the control SFRC beam

The mechanical behavior of the control beam B1.1-NC is analyzed based on the load-displacement curves, as shown in Fig.7 The solid line shows the mid-span displacement measured at the bottom face (denoted fb), while the dotted line presents the mid-span displacement measured at the tested beam’s neutral axis (denoted fn) These load-displacement curves could be divided into three main

103

Trang 8

stages At first, the stage OA represents the linear behavior of the SFRC beam until the first flexural

crack appears After the cracking load of approximately 80 kN, this beam begins to behave non-linear

in stage AB due to the formation of flexural cracks between two loading points, followed by shear

cracks from the support to the loading point In the next stage BC, the vertical displacement

con-tinues to increase quickly due to the opening of flexural cracks and shear cracks while maintaining

the applied load with a small variation The control beam reaches the maximum load (denoted PNCmax)

of 335.87 kN and the corresponding displacement measured at the neutral axis (denoted fnNC) of

5.88 mm Then, the concrete in the compression zone is crushed while the tensile longitudinal rebars

are yielding Besides, the shear crack CR2is significantly opened on one side of the beam, as shown

in Fig.8 At point C, the displacement fn is 18.68 mm, corresponding to 1/50 of the span, and the

tested beam is considered to have failed by flexural-shear mode This displacement is greater than

the critical value based on current design codes [6,25] Therefore, the SFRC beam exhibits a higher

ductility behavior instead of the sudden shear failure of traditional RC beams

9

the mid-span displacement measured at the tested beam's neutral axis (denoted f n)

These load-displacement curves could be divided into three main stages At first, the

stage OA represents the linear behavior of the SFRC beam until the first flexural crack

appears After the cracking load of approximately 80 kN, this beam begins to behave

non-linear in stage AB due to the formation of flexural cracks between two loading

points, followed by shear cracks from the support to the loading point In the next stage

BC, the vertical displacement continues to increase quickly due to the opening of

flexural cracks and shear cracks while maintaining the applied load with a small

variation The control beam reaches the maximum load (denoted maxNC

P ) of 335.87 kN and the corresponding displacement measured at the neutral axis (denoted NC

n

f ) of 5.88

mm Then, the concrete in the compression zone is crushed while the tensile

longitudinal rebars are yielding Besides, the shear crack CR 2 is significantly opened on

one side of the beam, as shown in Fig 8 At point C, the displacement f n is 18.68 mm,

corresponding to 1/50 of the span, and the tested beam is considered to have failed by

flexural-shear mode This displacement is greater than the critical value based on

current design codes [6], [25]] Therefore, the SFRC beam exhibits a higher ductility

behavior instead of the sudden shear failure of traditional RC beams

Fig 7 Load-displacement curves of the control beam B1.1-NC

In this study, the CMOD values of two shear cracks, CR 1 and CR 2 , were measured

using two displacement devices, as illustrated in Fig 5 The load-CMOD curves in Fig

8 show that two shear cracks do not occur under loading with the load less than 200 kN

while appearing several flexural cracks After that, shear crack CR 2 begins to propagate

diagonally from the support towards the loading point, while shear crack CR 1 can be

observed at the load of 250 kN As the beam reaches the maximum load, the width

(denoted w) of the shear crack CR1 grows to around 0.1 mm, while the shear crack CR 2

0 50 100

150

200

250

300

350

400

Displacement f (mm)

P - fb

P - fn

A

P - f b

P - f n

Figure 7 Load-displacement curves of the control

beam B1.1-NC

10

quickly expands to almost 0.5 mm in width, which is over to acceptable crack width according to EC 2 [25] At the end of the test, the shear crack CR 2 propagated to the compression zone and opened with a 0.8 mm width The cracking pattern due to loading was captured for the front and back sides of the control beam and presented in Fig 9

Fig 8 Load-CMOD curves of the control beam B1.1-NC

(a) Beam captured at the final state (b) Cracking pattern Fig 9 Failure mode and cracking pattern of the control beam B1.1-NC

3.3 Mechanical behavior of the corroded SFRC beams

Similar to the control beam, the mechanical behavior of corroded SFRC beams B2.1-C and B3.1-C is also evaluated based on the curves of displacement and load-CMOD, as shown in Figs 10 – 13 The load-displacement curves demonstrate that the performance of these corroded beams has deteriorated under the effect of steel reinforcement corrosion, but the mechanical behavior in stages from OA to BC is quite equivalent compared to the control beam At first, the corroded beams in stage OA operate a quasi-linear behavior since they had several corrosion-induced cracks At point A, the first flexural crack occurred at the load values of 130 and 135 kN for beams B2.1-C and B3.1-C, respectively Then, these beams begin to increase the vertical displacement in stage AB, which is described by reducing the slope of the

load-0 50 100 150 200 250 300 350 400

-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Crack width w (mm)

CR1 CR2

P - w CR1

P - w CR2

Figure 8 Load-CMOD curves of the control beam

B1.1-NC

In this study, the CMOD values of two shear cracks, CR1 and CR2, were measured using two displacement devices, as illustrated in Fig.5 The load-CMOD curves in Fig.8show that two shear

cracks do not occur under loading with the load less than 200 kN while appearing several flexural

cracks After that, shear crack CR2 begins to propagate diagonally from the support towards the

10

quickly expands to almost 0.5 mm in width, which is over to acceptable crack width

compression zone and opened with a 0.8 mm width The cracking pattern due to loading

was captured for the front and back sides of the control beam and presented in Fig 9

Fig 9 Failure mode and cracking pattern of the control beam B1.1-NC

Similar to the control beam, the mechanical behavior of corroded SFRC beams B2.1-C and B3.1-C is also evaluated based on the curves of displacement and

load-CMOD, as shown in Figs 10 – 13 The load-displacement curves demonstrate that the

performance of these corroded beams has deteriorated under the effect of steel

reinforcement corrosion, but the mechanical behavior in stages from OA to BC is quite

equivalent compared to the control beam At first, the corroded beams in stage OA

operate a quasi-linear behavior since they had several corrosion-induced cracks At

point A, the first flexural crack occurred at the load values of 130 and 135 kN for beams

B2.1-C and B3.1-C, respectively Then, these beams begin to increase the vertical

displacement in stage AB, which is described by reducing the slope of the

load-0 50 100 150 200 250 300 350 400

CR1 CR2

P - w CR1

P - w CR2

(a) Beam captured at the final state Journal of Science and Technology in Civil Engineering HUCE 2022 ISSN 1859-2996

10

quickly expands to almost 0.5 mm in width, which is over to acceptable crack width according to EC 2 [25] At the end of the test, the shear crack CR2 propagated to the compression zone and opened with a 0.8 mm width The cracking pattern due to loading was captured for the front and back sides of the control beam and presented in Fig 9

(a) Beam captured at the final state (b) Cracking pattern Fig 9 Failure mode and cracking pattern of the control beam B1.1-NC

Similar to the control beam, the mechanical behavior of corroded SFRC beams B2.1-C and B3.1-C is also evaluated based on the curves of displacement and load-CMOD, as shown in Figs 10 – 13 The load-displacement curves demonstrate that the performance of these corroded beams has deteriorated under the effect of steel reinforcement corrosion, but the mechanical behavior in stages from OA to BC is quite equivalent compared to the control beam At first, the corroded beams in stage OA operate a quasi-linear behavior since they had several corrosion-induced cracks At point A, the first flexural crack occurred at the load values of 130 and 135 kN for beams B2.1-C and B3.1-C, respectively Then, these beams begin to increase the vertical displacement in stage AB, which is described by reducing the slope of the

load-0 50 100 150 200 250 300 350 400

-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

CR1 CR2

P - w CR1

P - w CR2

(b) Cracking pattern Figure 9 Failure mode and cracking pattern of the control beam B1.1-NC

104

Trang 9

loading point, while shear crack CR1 can be observed at the load of 250 kN As the beam reaches

the maximum load, the width (denoted w) of the shear crack CR1grows to around 0.1 mm, while the

shear crack CR2quickly expands to almost 0.5 mm in width, which is over to acceptable crack width

according to EC 2 [25] At the end of the test, the shear crack CR2 propagated to the compression

zone and opened with a 0.8 mm width The cracking pattern due to loading was captured for the front

and back sides of the control beam and presented in Fig.9

Similar to the control beam, the mechanical behavior of corroded SFRC beams B2.1-C and

B3.1-C is also evaluated based on the curves of load-displacement and load-B3.1-CMOD, as shown in Figs

10–13 The load-displacement curves demonstrate that the performance of these corroded beams has

deteriorated under the effect of steel reinforcement corrosion, but the mechanical behavior in stages

from OA to BC is quite equivalent compared to the control beam At first, the corroded beams in

stage OA operate a quasi-linear behavior since they had several corrosion-induced cracks At point A,

the first flexural crack occurred at the load values of 130 and 135 kN for beams B2.1-C and

B3.1-C, respectively Then, these beams begin to increase the vertical displacement in stage AB, which is

described by reducing the slope of the load-displacement curves

11

displacement curves

Fig 10 Load-displacement curves of the corroded beam B2.1-C

In the next stage BC, the behavior of the two corroded beams is referred to in the

sustaining stage as the control beam The results show that beam B2.1-C has the

maximum load (denoted maxC

P ) of 226.58 kN at the corresponding displacement measured at the neutral axis (denoted C

n

f ) of 13.18 mm (Fig 10) Meanwhile, the obtained results of beam B3.1-C are maxC

P of 251.90 kN and C

n

f of 14.75 mm, as shown

in Fig 11 Therefore, the average maximum load of these corroded beams equals 239.24

kN On the other hand, when the applied load reaches 60% of the maximum load, the

shear crack CR 1 does not expand, while the shear crack CR 2 begins to open

continuously, as can be seen in Figs 12 and 13 In stage BC, the shear crack CR 2 grows

beyond 5 mm After that, the measured width may be inaccurate due to the separation

0 50

100

150

200

250

P - fb

P - fn

A

D

P - f b

P - f n

0 50

100

150

200

250

300

P-CV P-CVth

A

D

P - f b

P - f n

Figure 10 Load-displacement curves of the

corroded beam B2.1-C

11

displacement curves

In the next stage BC, the behavior of the two corroded beams is referred to in the sustaining stage as the control beam The results show that beam B2.1-C has the maximum load (denoted maxC

P ) of 226.58 kN at the corresponding displacement measured at the neutral axis (denoted C

n

f ) of 13.18 mm (Fig 10) Meanwhile, the obtained results of beam B3.1-C are maxC

P of 251.90 kN and C

n

f of 14.75 mm, as shown

in Fig 11 Therefore, the average maximum load of these corroded beams equals 239.24

kN On the other hand, when the applied load reaches 60% of the maximum load, the shear crack CR 1 does not expand, while the shear crack CR 2 begins to open continuously, as can be seen in Figs 12 and 13 In stage BC, the shear crack CR 2 grows beyond 5 mm After that, the measured width may be inaccurate due to the separation

0 50 100 150 200 250

P - fb

P - fn

A

D

P - f b

P - f n

0 50 100 150 200 250 300

P-CV P-CVth

A

D

P - f b

P - f n

Figure 11 Load-displacement curves of the

corroded beam B3.1-C

In the next stage BC, the behavior of the two corroded beams is referred to in the sustaining stage

as the control beam The results show that beam B2.1-C has the maximum load (denoted PCmax) of

226.58 kN at the corresponding displacement measured at the neutral axis (denoted fnC) of 13.18 mm

(Fig.10) Meanwhile, the obtained results of beam B3.1-C are PCmaxof 251.90 kN and fnCof 14.75 mm,

as shown in Fig.11 Therefore, the average maximum load of these corroded beams equals 239.24 kN

On the other hand, when the applied load reaches 60% of the maximum load, the shear crack CR1

does not expand, while the shear crack CR2begins to open continuously, as can be seen in Figs.12

and13 In stage BC, the shear crack CR2grows beyond 5 mm After that, the measured width may be

inaccurate due to the separation of the concrete cover near the support

In the stage CD, the corroded beams were fractured due to the loss of load-carrying capacity On

the corroded beam B2.1-C, the shear crack CR1propagates and intersects with the longitudinal crack

due to corrosion to become a web-shear crack, as shown in Fig 14 Besides, the shear crack CR2

continues to open with a significant width Due to the ductility of SFRC, both corroded beams have

relatively similar cracking patterns until failure In particular, during the experiment, beam B3.1-C

105

Trang 10

12

of the concrete cover near the support

Fig 12 Load-CMOD curves of the corroded beam B2.1-C

In the stage CD, the corroded beams were fractured due to the loss of load-carrying

capacity On the corroded beam B2.1-C, the shear crack CR 1 propagates and intersects

with the longitudinal crack due to corrosion to become a web-shear crack, as shown in

Fig 14 Besides, the shear crack CR 2 continues to open with a significant width Due to

the ductility of SFRC, both corroded beams have relatively similar cracking patterns

until failure In particular, during the experiment, beam B3.1-C suddenly failed at the

load of 200 kN owing to the corroded rebar rupture, as shown in Fig 15 After

demolition of surrounding concrete, ruptured steel rebar at the tension layer, named

R3-2, was measured to obtain a severe corrosion degree of 19.2%, the highest value among

all longitudinal steel rebars, as indicated in Table 4 Therefore, the final failure of this

0 50

100

150

200

250

-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

CR1 CR2

P - w CR1

P - w CR2

0 50

100

150

200

250

300

-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

CR1 CR2

P - w CR1

P - w CR2

Figure 12 Load-CMOD curves of the corroded

beam B2.1-C

12

of the concrete cover near the support

In the stage CD, the corroded beams were fractured due to the loss of load-carrying capacity On the corroded beam B2.1-C, the shear crack CR 1 propagates and intersects with the longitudinal crack due to corrosion to become a web-shear crack, as shown in Fig 14 Besides, the shear crack CR 2 continues to open with a significant width Due to the ductility of SFRC, both corroded beams have relatively similar cracking patterns until failure In particular, during the experiment, beam B3.1-C suddenly failed at the load of 200 kN owing to the corroded rebar rupture, as shown in Fig 15 After demolition of surrounding concrete, ruptured steel rebar at the tension layer, named

R3-2, was measured to obtain a severe corrosion degree of 19.2%, the highest value among all longitudinal steel rebars, as indicated in Table 4 Therefore, the final failure of this

0 50 100 150 200 250

-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

CR1 CR2

P - w CR1

P - w CR2

0 50 100 150 200 250 300

-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

CR1 CR2

P - w CR1

P - w CR2

Figure 13 Load-CMOD curves of the corroded

beam B3.1-C

suddenly failed at the load of 200 kN owing to the corroded rebar rupture, as shown in Fig 15

After demolition of surrounding concrete, ruptured steel rebar at the tension layer, named R3-2, was

measured to obtain a severe corrosion degree of 19.2%, the highest value among all longitudinal steel

rebars, as indicated in Table4 Therefore, the final failure of this beam is different compared to the

tested others

13

beam is different compared to the tested others

(a) Beam captured at the final state (b) Cracking pattern Fig 14 Failure mode and cracking pattern of the corroded beam B2.1-C

(a) Beam captured at the final state (b) Cracking pattern Fig 15 Failure mode and cracking pattern of the corroded beam B3.1-C Additionally, Figs 14 and 15 illustrate the failure mode and cracking patterns on two corroded beams Compared to the control beam, the corroded beams have two types

of cracks, including corrosion-induced cracks and cracks due to loading In the

beginning, the corrosion-induced cracks were horizontally distributed along the length

of the beam specimens After conducting the loading test, the flexural cracks began to

appear from the bottom of the beam These cracks tend to develop and intersect with

the corrosion-induced cracks as the load increases The failure is initiated by crushing

the concrete in the compression zone, while a flexural crack at the middle span in the

tension zone is opened with a significant width Besides, the shear cracks intersected

with the horizontal cracks due to corrosion and propagated quickly to the loading point

or crushed concrete zone The failure mode of the corroded beam B2.1-C is

shear-tension, which is characterized by the web-shear cracks that occurred due to the bond

loss between corroded longitudinal reinforcement and concrete Meanwhile, the

corroded beam B3.1-C failed by the shear-tension with the corroded rebar rupture

3.4 Comparisons between control beam and corroded beam

As shown in Fig 16, the corroded beams have a stiffness smaller than the control

(a) Beam captured at the final state Journal of Science and Technology in Civil Engineering HUCE 2022 ISSN 1859-2996

13

beam is different compared to the tested others

(a) Beam captured at the final state (b) Cracking pattern Fig 14 Failure mode and cracking pattern of the corroded beam B2.1-C

(a) Beam captured at the final state (b) Cracking pattern Fig 15 Failure mode and cracking pattern of the corroded beam B3.1-C Additionally, Figs 14 and 15 illustrate the failure mode and cracking patterns on two corroded beams Compared to the control beam, the corroded beams have two types

of cracks, including corrosion-induced cracks and cracks due to loading In the beginning, the corrosion-induced cracks were horizontally distributed along the length

of the beam specimens After conducting the loading test, the flexural cracks began to appear from the bottom of the beam These cracks tend to develop and intersect with the corrosion-induced cracks as the load increases The failure is initiated by crushing the concrete in the compression zone, while a flexural crack at the middle span in the tension zone is opened with a significant width Besides, the shear cracks intersected with the horizontal cracks due to corrosion and propagated quickly to the loading point

or crushed concrete zone The failure mode of the corroded beam B2.1-C is shear-tension, which is characterized by the web-shear cracks that occurred due to the bond loss between corroded longitudinal reinforcement and concrete Meanwhile, the corroded beam B3.1-C failed by the shear-tension with the corroded rebar rupture

3.4 Comparisons between control beam and corroded beam

As shown in Fig 16, the corroded beams have a stiffness smaller than the control

(b) Cracking pattern Figure 14 Failure mode and cracking pattern of the corroded beam B2.1-C

Additionally, Figs.14 and15 illustrate the failure mode and cracking patterns on two corroded

beams Compared to the control beam, the corroded beams have two types of cracks, including

corrosion-induced cracks and cracks due to loading In the beginning, the corrosion-induced cracks

were horizontally distributed along the length of the beam specimens After conducting the loading

test, the flexural cracks began to appear from the bottom of the beam These cracks tend to develop

and intersect with the corrosion-induced cracks as the load increases The failure is initiated by

crush-ing the concrete in the compression zone, while a flexural crack at the middle span in the tension zone

is opened with a significant width Besides, the shear cracks intersected with the horizontal cracks due

to corrosion and propagated quickly to the loading point or crushed concrete zone The failure mode

of the corroded beam B2.1-C is shear-tension, which is characterized by the web-shear cracks that

occurred due to the bond loss between corroded longitudinal reinforcement and concrete Meanwhile,

the corroded beam B3.1-C failed by the shear-tension with the corroded rebar rupture

106

Định dạng
Số trang	14
Dung lượng	4,93 MB